The miRNA Expression of Urinary Extracellular Vesicles in Patients With Gitelman Syndrome: The Role of hsa‐let‐7d‐3p

ABSTRACT

Gitelman syndrome (GS) is caused by an inactivating mutation in the SLC12A3, encoding the thiazide‐sensitive sodium chloride cotransporter (NCC), leading to salt wasting and electrolyte imbalance. Our aim is to identify microRNA (miRNA) expression and explore their role from urinary extracellular vesicles (uEVs) in GS patients. In this study, both uEVs from 23 genetically confirmed GS patients and renal biopsied tissues from another 3 GS patients were extracted for small RNA sequencing. Small RNA sequencing identified 358 miRNAs from uEVs and 652 miRNAs from renal biopsied tissues. Among differentially expressed miRNAs, 20 were upregulated and 23 downregulated in uEVs, while renal biopsied tissues showed 30 upregulated and 23 downregulated miRNAs. Four miRNAs (hsa‐let‐7d‐3p, hsa‐miR‐362‐5p, hsa‐miR‐30c‐5p, and hsa‐miR‐30b‐5p) overlapped from uEVs and renal tissues. In particular, the terms of the distinct upregulated hsa‐let‐7d‐3p target genes were related to ion transport and membrane depolarization, especially in neural precursor cell expressed, developmentally downregulated 4‐like (NEDD4L). Real‐time PCR of uEVs from another 11 GS patients confirmed significantly elevated hsa‐let‐7d‐3p compared to healthy controls. The decrease in the Nedd4l expression in the collecting duct was also confirmed in Ncc^S707X/S707X knock‐in mice. Dual luciferase assays further demonstrated that hsa‐let‐7d‐3p negatively regulated the expression of NEDD4L. These findings concluded that differentially expressed miRNAs could be identified from uEVs in GS patients, and hsa‐let‐7d‐3p, the only upregulated miRNA, negatively regulates NEDD4L expression in the collecting duct.

This study identifies differentially expressed miRNAs in urinary extracellular vesicles and kidney biopsies from Gitelman syndrome patients. In particular, hsa‐let‐7d‐3p is the only upregulated miRNA that negatively regulates NEDD4L, a key modulator of sodium transport. Functional validation in a GS mouse model and dual‐luciferase assays confirms its inhibitory role. These findings highlight uEVs as a non‐invasive tool for studying miRNA‐mediated renal transporter regulation, providing new insights into GS pathophysiology.

1. Introduction

Gitelman syndrome (GS) is an autosomal recessive renal tubular disease [1] that affects approximately 1–10 in 40 000 individuals [2]. GS is caused by mutations in the SLC12A3 gene (solute carrier family 12, member 3) [3], which is responsible for the sodium chloride cotransporter (NCC) and leads to salt loss, metabolic alkalosis, hypokalemia, hypomagnesemia, and hypocalciuria [4]. GS typically presents between adolescence and adulthood and can cause unspecific symptoms such as muscle weakness, fatigue, electrolyte imbalances, thirst, nocturia, and cramping. Since 1996, more than 500 different mutations have been identified in SLC12A3 [5, 6, 7]. The gold standard diagnosis of GS is based on a genetic test by direct Sanger sequencing or next‐generation sequencing, but the epigenetic regulation of microRNA (miRNA) in renal tubular transporters in GS remains unclear. In recent years, researchers have focused on the use of urinary extracellular vesicles (uEVs) proteins as a biomarker, particularly in inherited renal tubular diseases [8]. Our previous study has shown that uEVs could be a noninvasive tool to diagnose and evaluate the renal tubular transporter in patients with GS, showing an increased abundance of NCC and p‐NCC, as well as NHE3, NKCC2, ENaCβ, pendrin, ROMK, and Maxi‐K, similar to immunofluorescence of their kidney tissues [9]. However, the expression of miRNAs from uEVs in patients with GS remains unknown.

Extracellular vesicles (EVs) are membrane‐contained vesicles released by cells and transfer information to other cells, thereby influencing the recipient cell function. EVs are released by cells into extracellular space through membrane fusion and classified into three classes: microvesicles/microparticles/ectosomes, exosomes, and apoptotic bodies [10]. EVs can be isolated from body fluids such as saliva, blood, milk, and urine [11, 12]. The phospholipid bilayers of EVs protect their cargo from degradation by RNase or protease [13], making them a suitable biomarker and therapeutic delivery system. The biomolecules of uEVs include miRNA, mRNA, and proteins derived from cells and may provide signal transduction for cell‐to‐cell interaction. Among these biomolecules, miRNA is a short, endogenous, noncoding RNA molecule ranging from 21 to 25 bp [14], which plays an essential role in the regulation of normal embryonic and organ development [15] and is involved in many human diseases. By binding to the 3′ untranslated regions (UTRs) of the messenger RNA, miRNA inhibits gene transcription [16] and increases mRNA degradation, ultimately affecting gene expression. The miRNA is known to play a crucial role in renal development and physiological regulation, and its dysregulation can contribute to the severity of renal diseases [17, 18, 19, 20, 21, 22, 23, 24]. Isolation of uEVs miRNA has been used in various kidney‐associated diseases, such as diabetes (miR‐145) [25, 26, 27, 28], SLE (miR‐146a, miR‐26a) [29], IgA nephropathy (miR‐29c, miR‐146a, miR‐205) [30, 31], polycystic kidney disease and congenital abnormalities of the kidney and urinary tract (CAKUT) [32, 33]. However, the association between miRNAs in uEVs and renal tubular diseases remains unclear. It is hypothesized that the expression of these and other miRNAs may be altered in uEVs in patients with GS. Our aim is to study the differential expression of miRNAs in uEVs from patients with GS and to investigate the biological function of the regulation of renal tubular transporters.

The results to be reported indicated that the uEVs contained differentially expressed miRNAs such as hsa‐let‐7d‐3p, hsa‐miR‐362‐5p, hsa‐miR‐30c‐5p, and hsa‐miR‐30b‐5p that were validated by biopsied kidney tissues. Among them, hsa‐let‐7d‐3p was the only upregulated miRNA that negatively regulates neural precursor cell expressed, developmentally downregulated 4‐like (NEDD4L) as confirmed by the functional dual luciferase reporter assay. This study provides a novel understanding of the miRNA function of uEVs in GS.

2. Materials and Methods

2.1. Study Design

The study protocol was approved by the Ethics Committee for Human Studies of Tri‐Service General Hospital (TSGHIRB No. 2‐108‐05‐072, No. B202005077, No. C202105041). We collected urine samples from 23 healthy controls (HC) and 23 GS patients with definite SLC12A3 mutations for the isolation of uEVs. Clinical characteristics and laboratory examinations were collected and determined. Furthermore, renal biopsied tissues were obtained from three different patients with GS with definite SLC12A3 mutations (compound heterozygous mutation of c.1924C>T/c.2548 + 253C>T, compound heterozygous mutation of c.1670‐191C>T/c.2661_2747del, and homozygous mutation of c.2875_2876delAG/c.2875_2876delAG) who had long‐term severe hypokalemia refractory to aggressive potassium (K⁺) supplementation and significant proteinuria. Control kidney tissues were obtained from the normal part of the kidney in three patients with renal cell carcinoma.

2.2. Isolation and Validation of uEVs

The 40 mL urine mixture for each group was collected from patients and health controls who already added the cOmplete EDTA‐free protease inhibitor cocktail tablet (Roche, Mannheim, Germany) to prevent protein degradation. The samples were centrifuged at 17,000 × g for 10 min at 4°C to remove the cell and the cell debris. Then we used Vivaspin turbo 15 (Sartorius AG, Germany) at 2000 × g for 10 min at 4°C to concentrate the sample volume to 10 mL. Add the ExoQuick‐TC precipitation solution (System Biosciences, CA, USA) 2 mL to a concentrated sample of 10 mL and refrigerate at 4°C overnight. Finally, the samples were centrifuged at 1500 × g for 30 min at 4°C to precipitate the uEVs pellet. The characteristics of uEVs were validated by nanoparticle tracking analysis (NTA) (Malvern Panalytical, UK), transmission electron microscopy (TEM) HT7700 (Hitachi, Tokyo, Japan), and immunoblotting with antibodies to uEVs markers including NSE (ab79757, Abcam, Cambridge, UK), TSG101 (ab125011, Abcam, Cambridge, UK), CD9 (GTX55564, Genetex, HsinChu City) following the published methods [34].

2.3. Isolation of Small RNA From uEVs and Renal Biopsied Tissues

Small RNAs from uEVs and renal biopsy tissues were extracted using the miRNeasy micro kit (Qiagen, Hilden, Germany), following the manufacturer's protocol. The uEVs and renal biopsy tissues were homogenized in QIAzol reagent. The homogenized sample was placed into the binding column, and several wash steps finally eluted the small RNA. Concentration and purity were confirmed using the Qubit fluorometer (Invitrogen, Waltham, USA) and the Agilent Bioanalyzer 2100 instrument with the small RNA assay kit (Agilent, Santa Clara, CA, USA).

2.4. Small‐RNA Sequencing

1 ng of each small RNA sample was used for the construction of small RNA libraries following the QIAseq miRNA Library kit protocol. After the small RNA samples undergo 3′ and 5′ end ligation, reverse transcription was performed. The resulting cDNA was purified using gel beads and then subjected to index ligation. Unique molecular indexes (UMI) were used to join reads with the same amplification origin into UMI reads. After purification, the libraries were measured for concentration and length, diluted to the appropriate concentration, and mixed with samples for sequencing. Sequencing was performed on the Illumina NextSeq 550 platform (Illumina, San Diego, CA, USA) according to the manufacturer's instructions. Sequences were size‐sorted and reads with lengths between 16 and 36 bases were used for further analysis after removing adapters and low‐quality sequences. Redundant reads with multiple hits at different genomic locations were excluded from the dataset for downstream analysis. Data analysis was performed with GENEGLOBE online software (https://geneglobe.qiagen.com/pl/analyze). DESeq2 was also used to normalize the expression (UMI read counts, UMIs) of each miRNA in a sample by calculating a scaling factor (for detailed information on DESeq2, please refer to the DESeq2 manual: https://www.bioconductor.org/packages/3.3/bioc/vignettes/DESeq2/inst/doc/DESeq2.pdf). Generated FASTQ files and metadata have been deposited in the National Center for Biotechnology Information Gene Expression Omnibus (GEO Accession GSE282363).

2.5. Bioinformatics Analysis

MiRWalk (http://mirwalk.umm.uni‐heidelberg.de/) and TargetScan databases (https://www.targetscan.org/vert_80/) were used to predict miRNA target genes. To improve the accuracy of target gene prediction and reduce the rate of false positives, we selected a common intersection among the two databases as a filtering condition. Only miRNAs that could predict targets within at least two databases were further analyzed. Gene ontology terms enriched in the set of identified target genes whose transcripts changed relative to nonresponding genes were identified using DAVID (The Database for Annotation, Visualization, and Integrated Discovery v6.8).

2.6. Immunofluorescence Staining of Nedd4l in the Kidney of Ncc^{S707X /S707X} Knock‐In Mice

Kidney sections of mice were obtained from Ncc^S707X/S707X knock‐in mice, a model representing GS, which carries a nonsense mutation in the slc12a3 gene (Ser707X), corresponding to the human p.Ser710X mutation [35]. Animal studies were approved by the Institutional Animal Care and Use Committee of the National Defense Medical University (IACUC‐21–111). After removal of paraffin and rehydration of kidney tissue sections, slides were heated in EDTA buffer (ThermoFisher, Waltham, MA, USA) to recover antigen. To block endogenous peroxidase activity, sections were treated with UltraVision Hydrogen Peroxide Block (Thermo Fisher, Waltham, MA, USA) at room temperature, followed by incubation in UltraVision Protein Block (Thermo Fisher, Waltham, MA, USA). After washing with PBS containing 0.1% Tween 20 (J.T. Baker, Avantor, USA), the tissues were incubated with primary antibodies against Nedd4l (#4013, Cell Signaling Technology, Danvers, MA, USA) at 4°C overnight. Then, species‐specific secondary antibodies conjugated to Alexa Fluor fluorophores (Thermo Fisher Scientific, Waltham, MA, USA) were applied. Immunofluorescence images were captured using a Zeiss LSM 880 confocal microscope (Carl Zeiss, Oberkochen, Germany) to assess Nedd4l expression in kidney tissues.

2.7. Real‐Time PCR of the Identified miRNA, hsa‐let‐7d‐3p

We collected urine samples from 11 additional patients with GS and 11 healthy human controls as well as Ncc^S707X/S707X knock‐in and wild‐type (WT) mice (each n = 3). The uEVs and miRNA were isolated and extracted as in the method mentioned above. All RNA samples were subjected to cDNA synthesis using the miRCURY LNA miRNA PCR Assay system (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Customized hsa‐let‐7d‐3p oligonucleotides (5′‐CUAUACGACCUGCUGCCUUUCU‐3′) (Qiagen, Hilden, Germany) were used for real‐time PCR, with the UniSp6 serving as a reference miRNA. Real‐time PCR was performed on a QuantStudio 5 real‐time PCR system (Thermo Fisher, MA).

2.8. Dual‐Luciferase Reporter Assays

We used the pMIR‐REPORT system (Invitrogen, Waltham, MA, USA) to construct a plasmid containing the 1 kb NEDD4L 3′‐UTR with a 7‐bp miR‐hsa‐let‐7d‐3p miRNA binding site, as identified by the TargetScan database. We performed reverse direction PCR to construct a mutant plasmid with a 12‐bp deletion of the hsa‐let‐7d‐3p miRNA binding site. HEK293T cells were seeded in 24‐well plates and transfected with 100 ng/well of the constructed plasmid along with 100 nM/well of miRNA mimic or negative control miRNA. After 48 h of incubation, luciferase activity was measured using the Luc‐Screen Extended‐Glow Luciferase Reporter Gene Assay System (Thermo Fisher, Waltham, MA, USA). β‐Galactosidase activity was measured simultaneously using the Galacto‐Light Plus β‐Galactosidase Reporter Gene Assay System (Thermo Fisher, Waltham, MA, USA) for normalization. Luminescence was detected with the CLARIOstar Plus microplate reader (BMG Labtech, Ortenberg, Germany) with readings taken for 0.1 s per well.

2.9. Statistics

The urine biochemistry data are expressed as mean ± standard deviation (SD). For comparisons of miRNA expression (normalized UMIs) between GS patients and healthy controls, a Student's paired t‐test was performed to calculate the statistic p‐value. All statistical analyses were performed with Prism software (GraphPad, Boston, MA, USA).

3. Results

3.1. Clinical Characteristics of GS Patients

All GS patients (n = 23, male/female = 18/5) with an average age of 37.4 ± 11.6 years old were normotensive. The serum and urine biochemistries were shown in Table 1 with their serum K⁺ 2.5 ± 0.7 mmol/L with high urinary K⁺ excretion (urine K⁺/creatinine = 0.49 ± 0.27; transtubular K⁺ gradient = 9.58 ± 4.92), hypomagnesemia (serum Mg²⁺ = 1.6 ± 0.7 mg/dL) and hypocalciuria (urine Ca²⁺/creatinine = 0.031 ± 0.038 mmol/mmol). Their mutations included homozygous intronic mutation (n = 5) and compound heterozygous mutation (n = 18) in the SLC12A3 gene encoding NCC (Table S1).

TABLE 1.

Serum and urine biochemistries in patients with GS (n = 23).

	Mean ± SD
Serum
BUN (mg/dL) (7–25)	15.0 ± 4.8
Creatinine (mg/dL) (0.5–0.9)	0.9 ± 0.3
Sodium (mmol/L) (136–145)	136.9 ± 3.3
Potassium (mmol/L) (3.5–5.1)	2.6 ± 0.7
Chloride (mmol/L) (98–107)	97.7 ± 3.3
Total Calcium (mmol/L) (8.6–10.2)	9.8 ± 0.4
Magnesium (mg/dL) (1.7–2.55)	1.6 ± 0.7
Hematocrit (%) (38.0–47.0)	44.5 ± 6.1
Albumin (g/dL) (3.5–5.7)	4.4 ± 0.5
Urine
Creatinine (mg/dL)	70.2 ± 35.2
Sodium (mmol/L)	80.5 ± 44.8
Potassium (mmol/L)	30.0 ± 15.3
Chloride (mmol/L)	77.6 ± 37.9
Calcium (mg/dL)	1.8 ± 2.2
Magnesium (mg/dL)	5.4 ± 3.5
Osmolality (mOsm/kg H2O)	403.0 ± 171.2
Transtubular K+ gradient (TTKG)	9.6 ± 4.9
K+/Creatinine (mmol/L/mg/dL)	0.49 ± 0.27

3.2. Characterization of the uEVs

The isolated uEVs of 5 HCs were confirmed by immunoblotting of uEVs markers, including TSG101, CD9, and NSE (Figure 1A). The size distribution of the particles identified by NTA showed a size range of 30 to 100 nm. Their mean particle size and concentration were 115.3 ± 4.4 nm and 2.1 × 10¹¹ ± 7.7 × 10¹⁰ particles/mL, respectively (Figure 1B). Figure 1C shows a representative real‐time monitoring image of the nanoparticles tracked by the NS300. TEM also confirmed the quality of the uEVs (Figure 1D).

FIGURE 1.

Characteristics of urinary extracellular vesicles (uEVs). uEVs were confirmed by immunoblotting (TSG101, CD9, and NSE) (A), nanoparticle tracking analysis with an average size distribution of 115.3 nm (B) and real‐time monitoring image (C), and electron micrography (scale bar 100 nm) (D).

3.3. Small RNA Sequencing of uEVs and Biopsied Kidney Tissues

We identified 358 miRNAs (Data S2) from uEVs, and there were 20 upregulated and 23 downregulated miRNAs in patients with GS compared to HC (male/female = 20/3 with an average age of 34.9 ± 8.7 years old) based on both criteria of |Log₂(GS UMIs/HC UMIs)| greater than 1 and p‐value less than 0.05 (Table S2). Furthermore, we also identified 652 miRNAs (Data S3), and there were 30 upregulated and 23 downregulated miRNAs from renal biopsied tissues (Table S3). Integrating differentially expressed miRNAs between renal biopsied tissues, there were 4 codifferentially expressed miRNAs, including hsa‐let‐7d‐3p, hsa‐miR‐30b‐5p, hsa‐miR‐30c‐5p, and hsa‐miR‐362‐5p (Table 2). Among these miRNAs, hsa‐let‐7d‐3p was the only upregulated miRNA, while the other three miRNAs were downregulated miRNAs. A volcano plot of uEVs (Figure 2A) and renal biopsied tissues (Figure 2B) showed differentially expressed miRNAs.

TABLE 2.

Differential expression of miRNA expression in GS patients from both uEVs and biopsied tissues.

	uEVs		Tissues
	Log2 (GS UMIs/HC UMIs)	p	Log2 (GS UMIs/HC UMIs)	p
hsa‐let‐7d‐3p	1.27	0.006	2.62	0.020
hsa‐miR‐362‐5p	−1.49	0.013	−1.41	0.024
hsa‐miR‐30c‐5p	−1.66	0.023	−1.04	0.031
hsa‐miR‐30b‐5p	−4.61	0.039	−1.57	0.032

FIGURE 2.

Differential expression miRNAs. The volcano plot showed the differential expression miRNAs (red points) from urinary extracellular vesicles (uEVs) (n = 358) (A) and renal biopsy tissues (n = 652) (B) when comparing GS patients to the healthy control (HC) (Log₂[GS UMIs/HC UMIs]). p values as indicated, Student's paired t‐test.

3.4. Biological Process Terms of Target Genes From Differentially Expressed miRNAs

To investigate the biological functional roles of differentially expressed miRNAs, we used the miRWalk and TargetScan databases to predict target genes. Among the downregulated miRNAs of hsa‐miR‐30b‐5p and hsa‐miR‐30c‐5p, we identified 30 shared target genes. The biological process of gene ontology highlighted the terms of small RNA loading in RISC, unwinding of the secondary RNA structure, pre‐miRNA processing, miRNA‐mediated gene silencing, and regulation of the ERK1/ERK2 cascade (Table 3). Concerning the downregulated hsa‐miR‐362‐5p, we identified 38 shared target genes. The biological process of gene ontology revealed the terms of regulation of mRNA stability and the Wnt signaling pathway.

TABLE 3.

Statistically over‐represented gene ontology biological process terms in the list of overlapping target genes (n = 30) of hsa‐miR‐30b‐5p and hsa‐miR‐30c‐5p by TargetScan.

Gene ontology biological process term	Gene	Fold enrichment	p (Fisher exact)
miRNA loading onto RISC involved in gene silencing by miRNA	AGO4, AGO1	179.2	5.20E–05
Small RNA loading onto RISC	AGO4, AGO1	159.3	6.70E–05
RNA secondary structure unwinding	AGO4, AGO1	159.3	6.70E–05
Pre‐miRNA processing	AGO4, AGO1	102.4	1.70E–04
Production of miRNAs involved in gene silencing by miRNA	AGO4, AGO1	89.6	2.20E–04
Regulation of ERK1 and ERK2 cascade	LYN, ZDHHC17	51.2	6.90E–04

In hsa‐let‐7d‐3p, we identified 79 overlapped target genes (1889 target genes from miRWalk and 410 target genes from TargetScan). (Figure S1). The biological process revealed the terms of RNA splicing, membrane depolarization, negative regulation of the ERK1/ERK2 cascade, hydrogen ion transmembrane transport, selection of the mRNA splice site, and ion transmembrane transport (Table 4). Of note, NEDD4L was on the gene lists of two gene ontology biological process terms of “Regulation of membrane depolarization” and “Regulation of ion transmembrane transport.” In addition, NEDD4L has been reported to regulate epithelial sodium channel (ENaC) degradation. Therefore, we hypothesize that increased hsa‐let‐7d‐3p inhibits the NEDD4L expression and then decreases ENaC degradation and increases sodium reabsorption from ENaC. Our previous study has reported that the ENaC was significantly increased in Ncc^S707X/S707X knock‐in mice [35]. To further investigate the expression of hsa‐let‐7d‐3p and Nedd4l in Ncc^S707X/S707X knock‐in mice, we performed real‐time PCR of hsa‐let‐7d‐3p from Ncc^S707x/S707x uEVs and hsa‐let‐7d‐3p expression significantly increased compared to wild type (Figure 3A). Furthermore, the immunofluorescence staining for Nedd4l and Aqp2 (a marker of the collecting duct) in kidney tissues from the animal model (Ncc^S707X/S707X knock‐in mice) showed a marked reduction in Nedd4l expression (green) in CD of GS mice compared to wild‐type controls (Figure 3B,C).

TABLE 4.

Statistically over‐represented gene ontology biological process terms in the list of hsa‐let‐7d‐3p overlapping target genes (n = 79) by TargetScan and miRWalk.

Gene ontology biological process term	Gene	Fold enrichment	p (Fisher exact)
Regulation of RNA splicing	CELF1, GRSF1, PTBP2	9.9	0.0034
Regulation of membrane depolarization	NEDD4L, HCN1	53	0.00062
Native regulation of ERK1 and ERK2 cascade	DUSP3, CNKSR3, SPRY3	9.6	0.0038
Hydrogen ion transmembrane transport	SLC36A1, SLC15A2, SLC25A4	6.8	0.008
mRNA splice site selection	CELF1, PTBP2	25.3	0.0028
Regulation of ion transmembrane transport	KCND3, NEDD4L, HCN1	5.9	0.014

FIGURE 3.

Validation of increased hsa‐let‐7d‐3p and decreased Nedd4l expression in mice. (A) Increased hsa‐let‐7d‐3p from Ncc^S707x/S707x uEVs using real‐time PCR. (B) Decreased Nedd4l expression was confirmed in Ncc^S707X/S707X mice with Aqp2 used as a tubular marker of collecting duct (CD). (C) Three pairs of tissues from wild‐type mice and Ncc^S707x/S707x were shown.

3.5. Validation of hsa‐let‐7d‐3p Expression From uEVs From GS Patients and Its Regulation of NEDD4L

To validate the identified upregulated miRNA hsa‐let‐7d‐3p from uEVs, we collected uEVs from another 11 patients with GS and extracted miRNA for real‐time PCR. The results showed a significant increase in hsa‐let‐7d‐3p expression from uEVs in patients with GS compared to HC (Figure 4). The hsa‐let‐7d‐3p has been predicted to be directed directly to the NEDD4L 3′ UTR (Figure 5A). A dual luciferase assay was also performed for validation. Cells carrying the pMIR‐REPORT luciferase plasmid with the hsa‐let‐7d‐3p binding site showed reduced luciferase activity compared to the control group (p < 0.05). However, cells that carried the mutated plasmid of the hsa‐let‐7d‐3p binding site did not show any significant change in luciferase activity, regardless of whether mimic let‐7d‐3p miRNA was added (p > 0.05). These findings suggested that hsa‐let‐7d‐3p has an inhibitory effect on NEDD4L expression compared to mutated hsa‐let‐7d‐3p (Figure 5B).

FIGURE 4.

Real‐time PCR of hsa‐let‐7d‐3p from human uEVs. (A) Amplification plots of hsa‐let‐7d‐3p from real‐time PCR of uEVs samples from Gitelman syndrome patients (GS, red) and healthy controls (HC, blue) (GS, n = 11; HC, n = 11). ΔRn, Δ Rn value. (B) Relative quantification data of hsa‐let‐7d‐3p abundance in panel A. p values as indicated, unpaired t test. Error bars indicate SD.

FIGURE 5.

The miRNA hsa‐let‐7d‐3p targeting the NEDD4L mRNA through a targeting sequence located at 3′‐UTR. (A) The miRNA recognition element (MRE), located at position 3744–3750 of the NEDD4L 3′UTR (untranslated region), binds to the miRNA sequence. The binding coding (CD) sequence (CGUAUAAA) is paired with hsa‐let‐7d‐3p (GCAUAUC). This site was used in the luciferase assay with the pMIR‐REPORT system (B). Dual‐luciferase reporter assays were performed to test the interaction of hsa‐let‐7d‐3p and its targeting sequence in the NEDD4L 3′‐UTR using constructs containing the predicted targeting sequence (pMIR‐REPORT‐NEDD4L‐WT) and mutated targeting sequence (pMIR‐REPORT‐NEDD4L‐Mut) cloned into the 3′‐UTR of the reporter gene (B). Data represents three independent experiments with triplicate measurements. *Indicates p < 0.05.

4. Discussion

We hypothesized that the expression of uEVs miRNAs reflects regulatory mechanisms in GS, particularly those involving renal tubular transporters. Our study identifies four differentially expressed miRNAs—hsa‐let‐7d‐3p, hsa‐miR‐362‐5p, hsa‐miR‐30c‐5p, and hsa‐miR‐30b‐5p—in uEVs and biopsied kidneys from GS patients. Interestingly, hsa‐let‐7d‐3p was the only upregulated miRNA, its expression validated by uEVs real‐time PCR from other GS patients. The gene ontology analysis linked the target gene of hsa‐let‐7d‐3p to processes such as regulation of membrane depolarization and ion transmembrane transport. Among its target genes, NEDD4L, known to regulate ENaC degradation, has been hypothesized to be downregulated in response to increased urinary sodium flow due to an NCC defect in GS. Our dual luciferase assays confirmed that hsa‐let‐7d‐3p negatively regulated the expression of NEDD4L. These findings suggested that uEVs miRNAs not only provide biomarkers for GS but also provide information on the epigenetic regulation of renal tubular transporters.

Despite the limited studies of miRNA in GS, we confirmed that the changes in the represented renal tubular transporters of uEVs resembled the same expression of biopsied tissues from our previous study [9]. This study is the first to use uEVs small RNA sequencing in patients with GS that is also confirmed by human biopsied kidney tissues, and we identified co‐downregulated miRNAs (hsa‐miR‐30b‐5p, hsa‐miR‐30c‐5p, hsa‐miR‐362‐5p) and the only co‐upregulated hsa‐let‐7d‐3p. Among the predicted target genes of co‐downregulated miRNAs (hsa‐miR‐30b‐5p, hsa‐miR‐30c‐5p), the biological process of the terms of the gene ontology was associated with the regulation of miRNA and ERK1/2 cascades, while hsa‐let‐7d‐3p was associated with the negative regulation of ERK1/2 cascades, RNA splicing, and the regulation of ion transmembrane transport. Both mentioned in the up‐ and downregulation, the ERK1/2 signaling pathway has been shown to influence sodium transport in the kidney through its interactions with NEDD4L and ENaC in the collecting duct. ERK1/2 can phosphorylate NEDD4L, which in turn regulates ENaC activity by targeting it for ubiquitination and degradation. Dysregulation of this pathway could contribute to abnormal sodium handling, providing further information on the molecular mechanisms underlying GS [36].

In addition to the ERK1/2 signaling pathway, another gene ontology biological process term of the hsa‐let‐7d‐3p target genes was ion channel regulation that included the HCN1, KCND3, and NEDD4L genes. In HCN1, the hyperpolarization‐activated cationic and cyclic nucleotide (HCN) family is related to ammonia and K⁺ [37]; K⁺ voltage‐gated channel subfamily D member 3 (KCND3) was previously confirmed to be expressed in the brain and heart [38, 39], but its function in the kidney remains unknown. Of note, NEDD4L is essential to regulate ENaC endocytosis and lysosomal degradation and further modulate sodium balance [40, 41, 42]. Although hsa‐let‐7d‐3p has been implicated in the calcitonin gene‐related peptide pathway in migraine [43] and visceral hypersensitivity in irritable bowel syndrome with diarrhea [44], its role in the regulation of NEDD4L in the kidney has not been reported, especially in GS. The miRNAs such as miR‐192 and miR‐200b have been reported to regulate other renal transporters, including sodium transport machinery and Na⁺/H⁺ exchange regulatory factor‐1 [45], supporting the potential importance of hsa‐let‐7d‐3p in the expression of NEDD4L in the collecting ducts of GS patients.

NEDD4L has been reported to downregulate ENaC in collecting tubules, so loss of NEDD4L function enhances ENaC activity, leading to increased sodium reabsorption, consistent with the finding related to salt loss in GS [40, 46, 47]. We propose that upregulated hsa‐let‐7d‐3p in GS will negatively regulate the expression of NEDD4L in the collecting duct, leading to decreased ENaC degradation and increased sodium reabsorption to compensate for salt loss in GS. Furthermore, we also confirm the decrease in Nedd4l expression in the kidneys of Ncc^S707X/S707X knock‐in mice and analyze the interaction between hsa‐let‐7d‐3p and NEDD4L in our dual luciferase experiment, supporting the hypothesis that hsa‐let‐7d‐3p inhibits NEDD4L expression. This suggests that hsa‐let‐7d‐3p plays a critical role in the pathophysiology of GS by downregulating NEDD4L expression, likely contributing to abnormal sodium reabsorption by improving ENaC activity (Figure 6). These findings provide novel insights into the molecular mechanisms of GS and highlight the role of miRNA‐mediated regulation in GS.

FIGURE 6.

Proposed role of uEVs hsa‐let‐7d‐3p in regulating NEDD4L expression in Gitelman syndrome (GS). (A) Thiazide‐sensitive sodium chloride co‐transporter (NCC) and NEDD4L that promote ENaC degradation in distal tubules involve the balancing sodium reabsorption from ENaC in healthy control (HC). (B) In GS patients, a mutation in SLC12A3 reduces NCC activity in the distal convoluted tubule resulting in salt wasting. Increased uEVs hsa‐let‐7d‐3p will negatively regulate NEDD4L, that will reduce ENaC degradation, leading to reabsorption of sodium from ENaC.

Our study still has some limitations. First, the sample size of GS patients is relatively small and will probably affect statistical power because GS is a rare inherited renal tubular disease and not common in clinical practice, especially for validation experiments. Second, uEV miRNAs generally exhibit low expression levels, which may reduce the stability of multiple testing correction. Third, although we demonstrated how hsa‐let‐7d‐3p regulates NEDD4L in vitro, further in vivo studies are required to fully establish the clinical relevance of these interactions. Lastly, the absence of disease control groups, such as salt‐wasting patients or hypokalemia patients, limits our ability to determine whether these miRNA dysregulations are specific to GS. Future studies that incorporate these comparisons would provide a deeper understanding of the specificity of miRNA alterations in GS.

In conclusion, we identified differentially expressed miRNAs from uEVs and biopsied kidneys from GS and explored their roles in the ERK signaling pathway. We also evaluated how hsa‐let‐7d‐3p regulates NEDD4L that involves ENaC degradation in the collecting duct in response to salt wasting. Our findings suggested that uEVs miRNAs may provide insightful epigenetic regulation of renal tubular transporters.

Author Contributions

C.‐C.S., M.‐H.C., S.‐H.L., and C.‐T.C. conceptualized the study design and methodology. C.‐T.C., M.‐H.C., H.‐Y.C., M.‐H.T., Y.‐J.L., K.‐C.P., and C.‐C.S. performed the experiment, data collection, analysis, and interpretation of the results. C.‐T.C., M.‐H.C., Y.‐J.L., and S.‐H.L. wrote the initial draft of the manuscript, and C.‐C.S. and S.‐H.L. provided critical revisions for intellectual content. All authors reviewed and approved the final version of the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: fsb271192‐sup‐0001‐DataS1.pdf.

Data S2: 358 differentially expressed miRNAs from uEVs.

Data S3: 652 differentially expressed miRNAs from renal biopsied tissues.

Figure S1: Overlapped 79 hsa‐let‐7d‐3p target genes from Targetscan and miRWalk databases.

Table S1: Characteristics of SLC12A3 mutation among 23 GS patients.

Table S2: Differentially expressed miRNAs from urinary extracellular vesicles (total n = 23).

Table S3: Differentially expressed miRNAs from renal biopsied tissues (total n = 5).

Chen C.‐T., Lin S.‐H., Chen M.‐H., et al., “The miRNA Expression of Urinary Extracellular Vesicles in Patients With Gitelman Syndrome: The Role of hsa‐let‐7d‐3p,” The FASEB Journal 39, no. 21 (2025): e71192, 10.1096/fj.202501394RRR.

Data Availability Statement

The data underlying this article will be shared at reasonable request to the corresponding author.